DanWiggins

TheEar wrote: As woofers have to displace good amounts of air,and the longer the linear cone travel the better(to some extent).The Tumult is a prime example,very long linear throw and even when aproaching excursion limits its much more at home then most woofers at half max excursion. But... In midrange use,XBL may help.The gain may be audible at very high SPL. However in tweeter use where piston travel is minute the gains are more on paper then real audible gains. XBL^2 isn't about long stroke - it is about low distortion. You can gain longer stroke if desired, but it really arose from an R&D effort into driver distortion. Flat BL - REGARDLESS OF EXCURSION LEVELS - will lower THD. This is a prime benefit for any driver. As far as tweeters go, consider the dispersion aspect. Which has better off-axis dispersion - a 0.5" dome or a 1" dome? The 0.5" dome will have a 3 dB @ 45 degrees off axis frequency of 27 kHz; the 1" dome is rolling off at 13 kHz. For wide dispersion, a smaller dome is desirable. HOWEVER, that often means reduced SPL - you have 1/4 the Sd for generation of SPL. BUT, with a longer throw motor, you can gain that SPL back. SPL is displacement - Sd times throw. Reduce the Sd to gain dispersion, increase throw to gain back SPL. A 0.5" dome tweeter with 2mm of clean stroke will generate as much - or more - SPL as most 1" diameter domes. But with enhanced off axis response. Meaning the reflections, and the total power response will be much smoother overall. Extension is also a benefit. Say you want to stick with the same size dome - you're happy with your current Sd. And you are happy with the current SPL potential - the excursion is fine. With an XBL^2 motor you can typically cut the voice coil length by a factor of 2 or more, for the same stroke (that also implies cutting the Mms significantly, too). That quarters the inductance, meaning the natural roll-off of the driver has been raised. If your dome is clean to 30-40 kHz before breakup modes, now you can have extension out there. Is that important? Some would argue no, some would argue yes. But what IS important is now you can actually reach that point. Whether you want to or not is the designer's choice, but at least they now have that choice. There's a lot of benefit to an XBL^2 motor, other than just stroke. If you look at the dynamic THD from a driver (some of the iEC and AES test standards spell out how to measure dynamic, "musically based" THD), you'll find that, in terms of dynamic THD, a driver with XBL^2 and an Xmag of 12mm has the same dynamic THD as a standard overhung driver with an Xmag of 16mm. You gain significant reductions in distortion over the majority of the operating range, such that you can decrease the actual "linear" limits of throw, and still show dynamic THD reduction. I know it has been popular to look at "max SPL @ F @ 10% THD" as the guideline for performance of a driver. However, IMHO that is as valid as looking at cars at their torque and HP peaks only. To me, and many others, the actual shape of the torque curve is critical. I'd rather have a torque curve that was flat from 1000 RPM to 5250 RPM, and at 300 lb-ft over that window, than a torque curve that started at 100 lb-ft at 1000 RPM and peaked at 350 lb-ft at 5250 RPM. On paper, which engine is more powerful? On the strip, it will be a completely different result... XBL^2 is about reducing the distortion of the driver at all points. In typical overhung and underhung designs, THD steadily increases from 1mm of excursion and beyond. With an XBL^2 motor, the THD typically stays flat - VERY low, until you're 85-90% of the way to linear excursion. That is the big advantage. formica wrote: Note that I haven't read Dr. Klippel's white papers, but one would think that the relative importance of BL, Cms, and Le would vary depending on which property is governing the system?... or... in other words... BL would be more significant (affect distortion more) in a system which is subjected to large magnetic forces while Cms would be more significant in a rigid system subjected to small forces? I'm also unaware of the principals behind XBL, so don't be shy to step in here... but if it performs as advertised, I'm not denying the possibility of reducing distortion by improving BL linearity in a small driver... but rather it's overall significance (versus improving Cms linearity for example). You are correct about the relative importance of the system; however, for 99.5% of all cases, the box compliance dominates, not the driver's compliance, at least in the bass range. Thus the big influence on the system from the driver is the motor itself, BL. Above that, the compliance becomes much less of an issue. BL and Le start to dominate. Le dominates once you're up near the roll-off of the driver. So for the majority of the range - basically below the roll-off corner on the high end - BL tends to dominate. XBL^2 is described at http://www.adireaudio.com/tech_papers/xbl2_motors.htm - specifically the PDF download file http://www.adireaudio.com/Files/XBL2DetailsPaper.pdf describes how we use one implementation. It's really the concept of one or more voice coils transversing two or more magnetic gaps that have the direction of flux the same. Dan Wiggins Adire Audio

Hi all, perhaps I can explain a bit about how XBL^2, and linear BL in general - benefits mids and tweeters... In drivers, per Dr. Wolfgang Klippel, there are 3 sources of THD: BL nonlinearity - motor strength changes as the driver moves Cms nonlinearity - suspension stiffness changes as the driver moves Le nonlinearity - driver inductance changes as the driver moves These three sources are what generate THD when a driver moves. BL accounts for ~65% of the THD in a driver, Cms for ~25%, and Le for the last 10%. The big source of THD in a driver is from a nonlinear BL curve. Cms nonlinearity was solved in the 30s, with the introduction of the progressive and regressive spiders; they maintain a relatively constant Cms value over a wide range of motion. Likewise, Le nonlinearity was solved in the early 60s by Ejvind Skaaning, and the heavy use of copper in motors. BL nonlinearities, however, have been the difficult part for designers to tackle, and as the primary source of driver THD, has been the most important one to crack. In traditional overhung or underhung drivers, BL starts to change as soon as you start to move. This is why measured BL curves show consistent parabolic shapes - the B field integrated over the voice coil length is different for every position of the voice coil. As such, the driver begins distortion at extremely small motion, and continues to increase THD has driver excursion increases. Linearizing the BL - making the BL flat over excursion - eliminates the THD associated with the driver when operating in its linear range. The wider and flatter you can make the BL curve, the lower the overall THD of the motor. XBL^2 is applicable to mids and tweeters for a couple of reasons: 1. Lower distortion. As soon as the BL changes, THD increases. Keeping BL constant over range of motion - even if it's just a few mm - will help lower THD. Since most mids are overhung, they have a completely parabolic BL curve, and will exhibit increasing THD with just a few mm of excursion. 2. Longer stroke. This allows a smaller diameter driver to be used to generate the same SPL. SPL is about displacement of air. Displace X amount of air, receive Y dB of SPL. Giving a designer the option of moving to a 4" mid, rather than a 5.25" mid, is a serious advantage. Off-axis response is enhanced with the smaller diameter driver, and the usual tradeoff with SPl limitation is gone. 3. Wider bandwidth. Strange it sounds, to claim more stroke AND wider bandwidth. However, the primary limiter of bandwidth on the upper end is not mass, nor compliance, or BL. It's inductance. The easiest way to reduce inductance is to not create it in the first place. Since inductance goes as the square of the number of turns, shortening the voice coil by 30% can halve your inductance, and double your bandwidth. XBL^2 uses very short voice coils to achieve very long strokes; for example, we have a 4" prototype driver with a 6mm long voice coil that has 5.8mm of one way linear stroke. Typically you'd need a voice coil around 14mm to achieve that kind of stroke. We cut the length of the VC way down, meaning much lower inductance. 4. Lower mass. Moving mass, along with BL, sets your efficiency. The lower the mass, the more efficiency you gain. And with a shorter voice coil, you cut the moving mass considerably. The Fs also increases, however. But for many designs a high Fs isn't an issue; efficiency is more important. Additionally, it's always trivial to add mass to a driver - removing it is often much more difficult. Give a driver designer a lower starting baseline of moving mass and they'll love you. 5. Higher BL. XBL^2 motors keep more of the voice coil in the peak B of the field, meaning more BL over the stroke. Motor strength is a "single point" function; motor efficiency is the integral of the area under the BL curve. The more area under the curve, the more potential for work with a given input, meaning the higher the efficiency. Making the curve wider and flatter, and keeping the B value high, means more efficiency in the design, especially over motion. 6. Less compression. The number 1 source of compression in a driver is NOT thermal; it is BL. As the BL drops, the driver loses efficiency. This is INDEPENDENT of the power delivered - it is purely an excursion based function. Power compression can take seconds to build up in woofers, and even tens or hundreds of milliseconds (dozens of cycles) in mids. BL compression, however, happens instantaneously on the first cycle and every cycle. Keeping the BL up over the entire stroke eliminates this source of compression. 7. Higher BL, take 2. Because the voice coil is so short, you can run a narrower gap. For a given angular deflection, the amount of axial displacement of the end of the former is dictated by the length of the former. Using a short voice coil, you can decrease the axial displacement for a given angle of deflection. This means you can run a narrower/tighter gap, and still have the same resistance to scraping/rocking. Narrowing the gap will increase the B field, meaning more BL. There are several other advantages to using XBL^2 in woofers, midranges, and tweeters, but these are the primary ones that our licensees are benefiting from. Typically the lower THD, longer throw, lower inductance, and lower mass are the most appreciated reasons, with the others being a strong 'second tier' set. Thanks, Dan Wiggins Adire Audio